Journal
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
Volume 23, Issue 7, Pages 9518-9528Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TITS.2021.3136588
Keywords
Permutation-invariance; state representation; autonomous driving
Categories
Funding
- NSF China [52072213, U20A20334]
- Tsinghua University-Toyota Joint Research Center for AI Technology of Automated Vehicle
Ask authors/readers for more resources
In this paper, a new state representation method called encoding sum and concatenation (ESC) is proposed for decision-making in autonomous driving. Unlike existing methods, ESC can handle variable number of surrounding vehicles without pre-defined sorting rules. The proposed method uses a feature neural network to encode the feature of each surrounding vehicle and obtain a representation vector. Experiments show that using ESC representation improves the policy learning accuracy by 62.2% compared to fixed-permutation representation.
In this paper, we propose a new state representation method, called encoding sum and concatenation (ESC), to describe the environment observation for decision-making in autonomous driving. Unlike existing state representation methods, ESC is applicable to the situation where the number of surrounding vehicles is variable and eliminates the need for manually pre-designed sorting rules, leading to higher representation ability and generality. The proposed ESC method introduces a feature neural network (NN) to encode the real-valued feature of each surrounding vehicle into an encoding vector, and then adds these vectors up to obtain the representation vector of the set of surrounding vehicles. Then, a fixed-dimensional and permutation-invariance state representation can be obtained by concatenating the set representation with other variables, such as indicators of the ego vehicle and road. By introducing the sum-of-power mapping, this paper has further proved that the injectivity of the ESC state representation can be guaranteed if the output dimension of the feature NN is greater than the number of variables of all surrounding vehicles. This means that the ESC representation can be used to describe the environment and taken as the inputs of learning-based policy functions. Experiments demonstrate that compared with the fixed-permutation representation method, the policy learning accuracy based on ESC representation is improved by 62.2%.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available